A Class of Tests Based Jointly on Sub-Sample Minimum and Median

A Class of Tests Based Jointly on Sub-Sample Minimum and Median

Authors

  • Arun Kumar
  • Manish Goyal Department of Statistics, Post Graduate Government College, Chandigarh, INDIA
  • Narinder Kumar Department of Statistics, Panjab University, Chandigarh, INDIA

Keywords:

Nonparametric Tests, Bahadur Efficiency, Shift Parameter, Simulation Study, U-Statistic

Abstract

In Statistics, nonparametric tests play a very important role when the data sets are not normal. In this paper, we develop a class of nonparametric tests to compare location parameters of two populations. The proposed test statistic is based jointly on minimum and median of the sub-samples. The mean and variance of the test statistic are derived. The test statistic has asymptotic normality under some assumptions. The proposed class of the tests is compared with its existing competitor’s w.r.t. Pitman and Bahadur asymptotic relative efficiency. As an illustration, the proposed test is applied to a real life data set. We carried out the Monte Carlo simulation study to find the power and level of significance of the proposed test.

Downloads

Published

2021-12-01

How to Cite

1.
Kumar A, Goyal M, Kumar N. A Class of Tests Based Jointly on Sub-Sample Minimum and Median: A Class of Tests Based Jointly on Sub-Sample Minimum and Median. JKSA [Internet]. 2021 Dec. 1 [cited 2022 Dec. 6];32(1):1-17. Available from: https://ojs.ksa.org.in/index.php/JKSA/article/view/36

Issue

Section

Articles